Em Wed, 22 Jan 2014 21:00:24 +0000 Sean Young <sean@xxxxxxxx> escreveu: > On Wed, Jan 22, 2014 at 09:10:58PM +0200, Antti Seppälä wrote: > > On 22 January 2014 18:29, Sean Young <sean@xxxxxxxx> wrote: > > > On Wed, Jan 22, 2014 at 05:46:28PM +0200, Antti Seppälä wrote: > > >> On 21 January 2014 14:28, Sean Young <sean@xxxxxxxx> wrote: > > >> > On Mon, Jan 20, 2014 at 09:39:43PM +0200, Antti Seppälä wrote: > > >> >> This patch series introduces a simple sysfs file interface for reading > > >> >> and writing wakeup scancodes to rc drivers. > > >> >> > > >> >> This is an improved version of my previous patch for nuvoton-cir which > > >> >> did the same thing via module parameters. This is a more generic > > >> >> approach allowing other drivers to utilize the interface as well. > > >> >> > > >> >> I did not port winbond-cir to this method of wakeup scancode setting yet > > >> >> because I don't have the hardware to test it and I wanted first to get > > >> >> some comments about how the patch series looks. I did however write a > > >> >> simple support to read and write scancodes to rc-loopback module. > > >> > > > >> > Doesn't the nuvoton-cir driver need to know the IR protocol for wakeup? > > >> > > > >> > This is needed for winbond-cir; I guess this should be another sysfs > > >> > file, something like "wakeup_protocol". Even if the nuvoton can only > > >> > handle one IR protocol, maybe it should be exported (readonly) via > > >> > sysfs? > > >> > > > >> > I'm happy to help with a winbond-cir implementation; I have the hardware. > > >> > > > >> > > > >> > Sean > > >> > > >> Nuvoton-cir doesn't care about the IR protocol because the hardware > > >> compares raw IR pulse lengths and wakes the system if received pulse > > >> is within certain tolerance of the one pre-programmed to the HW. This > > >> approach is agnostic to the used IR protocol. > > > > > > Your patch talks about scancodes which is something entirely different. > > > This should be renamed to something better. > > > > > > > I agree that for the nuvoton my choice of wording (scancode) was a > > poor one. Perhaps wakeup_code would suit both drivers? > > > > > So with the nuvoton you program a set of pulses and spaces; with the > > > winbond you set the protocol and the scancode. I don't think there is > > > any shared code here. Maybe it's better to implement the wakeup > > > sysfs files in the drivers themselves rather than in rcdev, I guess that > > > depends on whether there are other devices that implement similar > > > functionality. > > > > > > > The code to be shared is the logic of creating, parsing and formatting > > the sysfs file. In the end the drivers are only interested in getting > > a bunch of values to write to the hardware. > > > > I was thinking about adding another file (wakeup_protocol sounds good) > > which would tell what semantics are used to interpret the contents of > > wakeup_code file (rc6, rc5, nec or raw). Would this be a decent > > solution? > > Good idea. I like it. Not sure if you saw it, but there's already another patchset proposing that, that got submitted before this changeset: https://patchwork.linuxtv.org/patch/21625/ > > Sean -- Cheers, Mauro -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
